Skip to content

Conversation

@AffectionateCurry
Copy link
Collaborator

Made sure timings were correct for level 3

@simonguozirui
Copy link
Collaborator

@AffectionateCurry were these the new updates for tensor size (Level 3) or have they already been incorporated?

@AffectionateCurry
Copy link
Collaborator Author

I timed all level 3 problems and increased/decreased problem sizes iteratively on a H100 until they fell in the range of 1-15ms which is also similar to what I did for level 1 and 2.

See this post:
https://scalingintelligence.stanford.edu/blogs/kernelbenchv01/

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants